Nature Biotechnology
22, 53 - 54 (2003)
Published online: 7 December 2003; | doi:10.1038/nbt922
Recurrent gain of chromosomes 17q and 12 in cultured human embryonic stem cellsJonathan S Draper1, 6, Kath Smith2, 6, Paul Gokhale1, Harry D Moore3, Edna Maltby2, Julie Johnson4, Lorraine Meisner4, Thomas P Zwaka5, James A Thomson5
& Peter W Andrews11 Department of Biomedical Science, University of Sheffield, Western Bank, Sheffield S10 2TN, UK. 2 North Trent Clinical Cytogenetics Service, Sheffield Children's Trust, Western Bank, Sheffield S10 2TH, UK. 3 Section of Reproductive and Developmental Medicine, University of Sheffield, Royal Hallamshire Hospital, Jessop Wing, Sheffield S10 2SF, UK. 4 Cytogenetic Laboratory at Wisconsin State Laboratory of Hygiene, 465 Henry Mall, Madison, Wisconsin 53706, USA. 5 National Primate Research Center and the Department of Anatomy, University of Wisconsin−Madison Medical School, 1200 Capital Court, Madison, Wisconsin 53715, USA. 6 These authors contributed equally to this work.
Correspondence should be addressed to Peter W Andrews p.w.andrews@sheffield.ac.uk.We have observed karyotypic changes involving the gain of chromosome 17q in three independent human embryonic stem (hES) cell lines on five independent occasions. A gain of chromosome 12 was seen occasionally. This implies that increased dosage of chromosome 17q and 12 gene(s) provides a selective advantage for the propagation of undifferentiated hES cells. These observations are instructive for the future application of hES cells in transplantation therapies in which the use of aneuploid cells could be detrimental.
Human embryonic stem cells have been cultured for extended periods while retaining a diploid karyotype1,
2,
3. They are strikingly similar to human embryonal carcinoma (hEC) cells, the stem cells of teratocarcinomas, in that undifferentiated hEC and hES cells display comparable expression profiles of genes and antigens4. However, hEC cells are typically aneuploid, showing distinctive chromosomal abnormalities including the gain of chromosome 17q and the presence of one or more isochromosomes 12p5,
6,
7. Long-term culture of mouse ES cells can lead to a decrease in pluripotency and the gain of distinct chromosomal abnormalities8. Here we show that similar chromosomal changes, which resemble those observed in hEC cells from testicular cancer, can occur in hES cells.
A culture of hES cell line H7 (designated H7.S0; karyotype 46,XX) was maintained in Sheffield for several months (see Supplementary Methods online). The cells retained an undifferentiated hES cell phenotype (SSEA3+, TRA-1-60+, Oct4+, Sox2+)4,
9, an ability to differentiate (judged by differentiation in vitro from embryoid bodies with the formation of cells such as neurons and beating cardiac muscle cells) and a normal karyotype. However, all the cells of a subline, H7.S6, defrosted from an early freezing of H7.S0, had acquired chromosomal changes by passage 60 (about 6 months): the cells all had a 46,XX, der(6)t(6;17)(q27;q1) karyotype. These cells expressed markers of undifferentiated hES cells and retained an ability to differentiate in culture. The translocation involved gain of the complete long arm of chromosome 17, translocated to 6q, without apparent loss of chromosome 6q or 17q material; the cells were trisomic for 17q (Fig. 1). A subgroup of cells also had trisomy 12.
 | | Figure 1. Karyotypic changes affecting chromosomes 17 and 12 in sublines of H7 hES cells. | | |  | (a) The relationship between the different cultures of H7 and H14 cells that were maintained in Sheffield. H7 was originally cultured from a vial frozen in passage 15 at the University of Wisconsin. Cultures were then provided with an 'S' number to designate subsequent passages in Sheffield. Subcultures derived from subsequent freezings were designated by the 'S' number the cells had reached at the time of freezing. Different cultures were frozen and karyotyped as shown (see Supplementary Figure 1 online). A minimum of 20 metaphase spreads was analyzed for all karyo-types. The latter two H7.S14 karyotypes were confirmed by analyzing 200 interphases with FISH. (b) The derivative chromosome, der(6)t(6;17), present in the subline of H7 showing, respectively, the chromosome stained by G-banding, M-FISH for chromosome 17q marker, subtelomere probes for chromosome 6 and locus-specific probes (HER2, also known as ERBB2) for chromosome 17q. These data indicate the presence of a complete extra copy of 17q, translocated to chromosome 6, apparently without loss of chromosome 6 material. (c) An isochromosome 12p present in a clonal subline of the H7 cells adapted to culture on Matrigel in the absence of feeders or feeder-conditioned medium (subline H7.S6-5ANF), shown by G-banding and by FISH using a 12p subtelomeric probe. These cells retained the der(6)t(6;17)(q27;q1) chromosome.
 Full Figure and legend (50K) |
| |
Another diploid H7 subculture (H7.S9) was reestablished from a different early freezing. After four months, all the cells had become trisomic for the whole of chromosome 17. Over the same period, we studied a different hES cell line, H14 (designated H14.S0; 46,XY). This line was diploid, but a subsequent culture from a frozen stock, H14.S9, also became trisomic for chromosome 17. To further explore the recurrent gain of chromosome 17, we established yet another culture of H7 cells (H7.S14) from another early freezing. Interphase fluorescence in situ hybridization (FISH) analysis did not detect any cells with trisomy 17. After 22 passages (2 months), karyotyping revealed trisomy 17; interphase FISH indicated trisomy 17 in 76% of the cells and, after an additional 17 passages, in 95% of the cells.
In Sheffield, a gain of chromosome 17q was common to all hES cells in which chromosomal changes were detected, arising on four independent occasions in two different hES cell lines. A gain of chromosome 12 was also seen, but only in subpopulations of cells; no other consistent karyotypic changes were noted. Because the differentiated derivatives of ES and EC cells have limited growth potential, genetic changes that promote stem cell self-renewal at the expense of differentiation are likely to provide strong selective growth advantages for the variant sublines in which they occur. Thus, increased dosage of chromosome 17q gene(s) may provide hES cells with some strong advantage for maintenance in vitro or may inhibit apoptosis, and the evolution of hES cells in culture may parallel that of hEC cells in tumors. Indeed, when a subline of H7 carrying the t(6;17) translocation was cultured on Matrigel without fibroblasts or fibroblast-conditioned medium for over 25 passages, a clone of these feeder-independent H7 cells was found to carry an isochromosome, i(12p), the characteristic karyotypic marker of hEC cells (Fig. 1).
The original reports of hES cells indicated a stable diploid karyotype1. However, recently we have also seen karyotypic changes in H1 and H14 hES cell lines at the University of Wisconsin (Table 1). Mostly these changes involve gain of chromosome 12, although in one case amplification of part of chromosome 17q, together with a range of other changes, was seen. Why gain of chromosome 17 predominates in one laboratory and chromosome 12 in the other is unclear. Although initial culture protocols were, in principle, the same, the karyotypic changes observed in the Wisconsin cells were obtained after clonal selection or efforts to culture cells in the absence of feeders. Whether similar changes will be identified in other lines and other laboratories remains to be established.
| | Table 1. Examples of karyotypic changes in hES cells seen at the University of Wisconsin | | |  |
Full Table |
| |
The mechanism by which gain of chromosome 17q confers proliferative advantage is a matter of conjecture. In germ cell tumors, overexpression of GRB7, located on chromosome 17q, is a common feature of hEC cells7. Gain of 17q is also associated with neuroblastoma10, and genes implicated in apoptosis and differentiation are sited on 17q; for example, Survivin, encoding an inhibitor of apoptosis, lies at 17q25 (ref. 11). STAT3 and GRB2, the homologs of which regulate self-renewal and differentiation in mouse ES cells12, are encoded by human 17q. With respect to chromosome 12, Nanog, a master controller of pluripotency in murine ES cells13, lies on the boundary of the minimal 12p amplicon associated with hEC cells14. Karyotypic change occurs in mouse ES cells, correlating with a reduced ability to colonize the germ line in chimeric mice8. These changes commonly affect chromosome 8, but also occasionally chromosome 11, which is partly syntenic with human chromosome 17q8,
15.
Apart from clues they offer regarding the genetic control of stem cell self-renewal, our observations of chromosomal changes in hES cells suggest that caution is warranted when designing culture conditions, and especially feeder-free conditions, for the cells; in vitro evolution may select for adaptive genetic changes. The occurrence and potential detrimental effects of such karyotypic changes will need to be considered in the development of hES cell−based transplantation therapies.
Note: Supplementary information is available on the Nature Biotechnology website.
Received 9 July 2003; Accepted 20 October 2003; Published online: 7 December 2003.
REFERENCES
- Amit, M. et al. Dev. Biol. 227, 271–278 (2000).
- Reubinoff, B.E., Pera, M.F., Fong, C.Y., Trounson, A. & Bongso, A. Nat. Biotechnol. 18, 399–404 (2000).
- Thomson, J.A. et al. Science 282, 1145–1147 (1998).
- Draper, J.S., Pigott, C., Thomson, J.A. & Andrews, P.W. J. Anat. 200, 249–258 (2002).
- Atkin, N.B. & Baker, M.C. Lancet 2, 1349 (1982).
- Rodriguez, E. et al. Genes Chromosomes Cancer 8, 230–236 (1993).
- Skotheim, R.I. et al. Cancer Res. 62, 2359–2364 (2002).
- Longo, L., Bygrave, A., Grosveld, F.G. & Pandolfi, P.P. Transgenic Res. 6, 321–328 (1997).
- Henderson, J.K. et al. Stem Cells 20, 329–337 (2002).
- Westermann, F. & Schwab, M. Cancer Lett. 184, 127–147 (2002).
- Chiou, S.K., Jones, M.K. & Tarnawski, A.S. Med. Sci. Monit. 9, PI25–29 (2003).
- Burdon, T., Smith, A. & Savatier, P. Trends Cell Biol. 12, 432–438 (2002).
- Chambers, I. et al. Cell 113, 643–655 (2003).
- Mostert, M.C. et al. Oncogene 16, 2617–2627 (1998).
- Liu, X. et al. Dev. Dyn. 209, 85–91 (1997).
Acknowledgments
This work was supported in part by grants from the Biotechnology and Biological Sciences Research Council, Wellcome Trust and Yorkshire Cancer Research. We thank Henry Yuen and the Oskar Rennebohm Foundation for their gifts to the University of Wisconsin Foundation that helped support this work.
Competing interests statement:
The authors declare competing financial interests. |
